CovTransformer:用于SARS-CoV-2血统频率预测的变压器模型
Yinan Feng1,2, Emma E Goldberg2, Michael Kupperman2,3
1Earth and Environmental Sciences Division, Los Alamos National Laboratory, Los Alamos, NM, United States.
Virus evolution
|December 11, 2024
概括
一个新的机器学习模型CovTransformer准确地预测了两个月前的SARS-CoV-2血统频率. 这种基于变压器的方法超越了现有的流行病监测和识别新兴变异的方法.
科学领域:
- 病毒学 病毒学
- 计算生物学 计算生物学
- 流行病学 流行病学
背景情况:
- 数以百计的SARS-CoV-2血统在全球传播,因此需要准确预测血统频率.
- 预测血统主导对于理解未来变种的病原性和免疫逃脱至关重要.
研究的目的:
- 为SARS-CoV-2血统频率预测开发可靠的机器学习模型.
- 为了解决传统的回归方法的局限性,由于噪音和偏见的血统数据.
主要方法:
- 开发了基于变压器架构的机器学习模型CovTransformer.
- 在英国和美国的SARS-CoV-2血统数据上训练和测试模型,然后评估对其他国家和美国各州的概括.
- 将CovTransformer的性能与Nextstrain中使用的多项回归模型进行了比较.
主要成果:
- CovTransformer准确地预测了全球和美国州级的未来两个月的血统频率.
- 该模型的表现明显优于Nextstrain多项式回归模型.
- 追溯分析显示,CovTransformer可以提前平均7周识别主导血统.
结论:
- 变压器模型为准确的SARS-CoV-2预测和流行病监测提供了一个有希望的方法.
- CovTransformer提供了一个强大的工具,用于识别快速扩展的SARS-CoV-2血统.
- 这一进步有助于积极研究变异性致病性和免疫逃脱.
更多相关视频
相关概念视频
Determination of Expected Frequency
2.1K
Suppose one wants to test independence between the two variables of a contingency table. The values in the table constitute the observed frequencies of the dataset. But how does one determine the expected frequency of the dataset? One of the important assumptions is that the two variables are independent, which means the variables do not influence each other. For independent variables, the statistical probability of any event involving both variables is calculated by multiplying the individual...
2.1K
Relative Frequency Distribution
10.5K
A relative frequency distribution is the proportion or fraction of times a value occurs in a data set. To find the relative frequencies, one can divide each frequency by the total number of data points in the sample. It is very similar to a regular frequency distribution, except that instead of reporting how many data values fall in a class, a relative frequency distribution reports the fraction of data values that fall in a class. These fractions or proportions are called relative frequencies...
10.5K
Cumulative Frequency Distribution
6.8K
A cumulative frequency distribution is another type of frequency distribution. Instead of reporting how many data values fall in some classes, it reports how many data values are contained in either that class or any class to its left. Technically, it means the sum of frequencies of the class and all the classes below it in a frequency distribution. A cumulative frequency is calculated by adding the frequency of each class lower than the corresponding class interval or category. In general, a...
6.8K
Construction of Frequency Distribution
7.6K
A frequency distribution table can be constructed using the steps given below.
First, make a table with two columns—one with the title of the data that needs to be organized, and the other column for frequency. [Draw a third column for tally marks if needed]. Then, take a look at the items given in the data set and decide if an ungrouped frequency distribution table or a grouped frequency distribution table would be more suitable. If there are large sets of different values, then it is...
First, make a table with two columns—one with the title of the data that needs to be organized, and the other column for frequency. [Draw a third column for tally marks if needed]. Then, take a look at the items given in the data set and decide if an ungrouped frequency distribution table or a grouped frequency distribution table would be more suitable. If there are large sets of different values, then it is...
7.6K
Mutation, Gene Flow, and Genetic Drift
58.0K
In a population that is not at Hardy-Weinberg equilibrium, the frequency of alleles changes over time. Therefore, any deviations from the five conditions of Hardy-Weinberg equilibrium can alter the genetic variation of a given population. Conditions that change the genetic variability of a population include mutations, natural selection, non-random mating, gene flow, and genetic drift (small population size).
58.0K
Frequency-dependent Selection
21.8K
When the fitness of a trait is influenced by how common it is (i.e., its frequency) relative to different traits within a population, this is referred to as frequency-dependent selection. Frequency-dependent selection may occur between species or within a single species. This type of selection can either be positive—with more common phenotypes having higher fitness—or negative, with rarer phenotypes conferring increased fitness.
21.8K


